Inferring a Personalized Next Point-of-Interest Recommendation Model with Latent Behavior Patterns

Authors: Jing He, Xin Li, Lejian Liao, Dandan Song, William Cheung

AAAI 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments on two large-scale LBSNs datasets demonstrate the significant improvements of our model over several state-of-the-art methods.
Researcher Affiliation Academia 1BJ ER Center of HVLIP&CC, School of Comp. Sci., Beijing Institute of Technology, Beijing, China 2Department of Computer Science, Hong Kong Baptist University , Hong Kong, China {skyhejing, xinli, liaolj, sdd}@bit.edu.cn, william@comp.hkbu.edu.hk
Pseudocode Yes Algorithm 1 Our Proposed Methodology
Open Source Code No The paper does not provide any statement regarding the release of source code or a link to a code repository.
Open Datasets Yes We choose two large-scale datasets from real-world LBSNs, Foursquare and Gowalla, to conduct the experiments. Foursquare check-in data is within Los Angeles, provided by (Bao, Zheng, and Mokbel 2012), while Gowalla dataset is from (Cheng et al. 2012) with a complete snapshot.
Dataset Splits No The paper states: 'For other parameters, we tune them in the training sets to find the optimal values, and subsequently use them in the test set.' While this implies a form of validation for parameter tuning, it does not explicitly define a 'validation' dataset split with specific percentages or counts, or refer to a standard validation split.
Hardware Specification No The paper does not provide any specific details about the hardware (e.g., CPU, GPU, memory, cloud services) used to run the experiments.
Software Dependencies No The paper describes the algorithms and models used (e.g., BPR, EM) but does not provide details on specific software dependencies, programming languages, or their version numbers used for implementation.
Experiment Setup Yes We set λΘ to be 1 for both FPMCLR and our proposed model. The empirical settings of the number of latent behavior patterns are 4 and 6 for Gowalla dataset and Foursquare dataset, respectively. For other parameters, we tune them in the training sets to find the optimal values, and subsequently use them in the test set.